Mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-02 17:22:31 +01:00
Code Issues Releases Activity

[kaltura] fix extraction error for videos from multiple kaltura servers

Browse Source
This commit is contained in:
Remita Amine 2016-07-04 21:34:27 +01:00
parent dafafe7cf1
commit d9163ae3b6
1 changed files with 10 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
youtube_dl/extractor/kaltura.py
View File

@ -91,14 +91,15 @@ def _extract_url(webpage):
if mobj: if mobj:
embed_info = mobj.groupdict() embed_info = mobj.groupdict()
url = 'kaltura:%(partner_id)s:%(id)s' % embed_info url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
escaped_pid = re.escape(embed_info['partner_id'])
service_url = re.search( service_url = re.search(
'<script[^>]+src=(?:["\'])((?:https?:)?//.+?)/p/%(partner_id)s/sp/%(partner_id)s00/embedIframeJs' % embed_info, r'<script[^>]+src=["\']((?:https?:)?//.+?)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid),
webpage) webpage)
if service_url: if service_url:
url = smuggle_url(url, {'service_url': service_url.group(1)}) url = smuggle_url(url, {'service_url': service_url.group(1)})
return url return url
def _kaltura_api_call(self, video_id, actions, *args, **kwargs): def _kaltura_api_call(self, video_id, actions, service_url=None, *args, **kwargs):
params = actions[0] params = actions[0]
if len(actions) > 1: if len(actions) > 1:
for i, a in enumerate(actions[1:], start=1): for i, a in enumerate(actions[1:], start=1):
@ -106,7 +107,7 @@ def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
params['%d:%s' % (i, k)] = v params['%d:%s' % (i, k)] = v
data = self._download_json( data = self._download_json(
self._SERVICE_URL + self._SERVICE_BASE, (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
video_id, query=params, *args, **kwargs) video_id, query=params, *args, **kwargs)
status = data if len(actions) == 1 else data[0] status = data if len(actions) == 1 else data[0]
@ -116,7 +117,7 @@ def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
return data return data
def _get_kaltura_signature(self, video_id, partner_id): def _get_kaltura_signature(self, video_id, partner_id, service_url=None):
actions = [{ actions = [{
'apiVersion': '3.1', 'apiVersion': '3.1',
'expiry': 86400, 'expiry': 86400,
@ -126,10 +127,10 @@ def _get_kaltura_signature(self, video_id, partner_id):
'widgetId': '_%s' % partner_id, 'widgetId': '_%s' % partner_id,
}] }]
return self._kaltura_api_call( return self._kaltura_api_call(
video_id, actions, note='Downloading Kaltura signature')['ks'] video_id, actions, service_url, note='Downloading Kaltura signature')['ks']
def _get_video_info(self, video_id, partner_id): def _get_video_info(self, video_id, partner_id, service_url=None):
signature = self._get_kaltura_signature(video_id, partner_id) signature = self._get_kaltura_signature(video_id, partner_id, service_url)
actions = [ actions = [
{ {
'action': 'null', 'action': 'null',
@ -152,19 +153,16 @@ def _get_video_info(self, video_id, partner_id):
}, },
] ]
return self._kaltura_api_call( return self._kaltura_api_call(
video_id, actions, note='Downloading video info JSON') video_id, actions, service_url, note='Downloading video info JSON')
def _real_extract(self, url): def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {}) url, smuggled_data = unsmuggle_url(url, {})
service_url = smuggled_data.get('service_url')
if service_url:
self._SERVICE_URL = service_url
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
partner_id, entry_id = mobj.group('partner_id', 'id') partner_id, entry_id = mobj.group('partner_id', 'id')
ks = None ks = None
if partner_id and entry_id: if partner_id and entry_id:
info, flavor_assets = self._get_video_info(entry_id, partner_id) info, flavor_assets = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'))
else: else:
path, query = mobj.group('path', 'query') path, query = mobj.group('path', 'query')
if not path and not query: if not path and not query: